Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 4000 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.4411. Min Range: -0.4411

Max Act: 0.4411. Min Act: -0.0000

Data Index: 957209 (C4 (Web Text))

Max Activating Token Index: 543

Text #1

Max Range: 0.4411. Min Range: -0.4411

Max Act: 0.3799. Min Act: -0.0001

Data Index: 523753 (C4 (Web Text))

Max Activating Token Index: 1003

Text #2

Max Range: 0.4411. Min Range: -0.4411

Max Act: 0.3791. Min Act: -0.0001

Data Index: 298948 (C4 (Web Text))

Max Activating Token Index: 573

Text #3

Max Range: 0.4411. Min Range: -0.4411

Max Act: 0.3333. Min Act: -0.0001

Data Index: 439317 (C4 (Web Text))

Max Activating Token Index: 589

Text #4

Max Range: 0.4411. Min Range: -0.4411

Max Act: 0.3420. Min Act: 0.0001

Data Index: 548865 (C4 (Web Text))

Max Activating Token Index: 444

Text #5

Max Range: 0.4411. Min Range: -0.4411

Max Act: 0.3310. Min Act: 0.0000

Data Index: 527164 (C4 (Web Text))

Max Activating Token Index: 593

Text #6

Max Range: 0.4411. Min Range: -0.4411

Max Act: 0.3357. Min Act: -0.0001

Data Index: 735916 (C4 (Web Text))

Max Activating Token Index: 217

Text #7

Max Range: 0.4411. Min Range: -0.4411

Max Act: 0.3220. Min Act: -0.0000

Data Index: 734832 (C4 (Web Text))

Max Activating Token Index: 824

Text #8

Max Range: 0.4411. Min Range: -0.4411

Max Act: 0.3329. Min Act: -0.0000

Data Index: 467861 (C4 (Web Text))

Max Activating Token Index: 376

Text #9

Max Range: 0.4411. Min Range: -0.4411

Max Act: 0.3232. Min Act: 0.0001

Data Index: 1073441 (C4 (Web Text))

Max Activating Token Index: 250

Text #10

Max Range: 0.4411. Min Range: -0.4411

Max Act: 0.3328. Min Act: 0.0000

Data Index: 36220 (C4 (Web Text))

Max Activating Token Index: 677

Text #11

Max Range: 0.4411. Min Range: -0.4411

Max Act: 0.3329. Min Act: -0.0001

Data Index: 602277 (C4 (Web Text))

Max Activating Token Index: 212

Text #12

Max Range: 0.4411. Min Range: -0.4411

Max Act: 0.3279. Min Act: -0.0000

Data Index: 714415 (C4 (Web Text))

Max Activating Token Index: 152

Text #13

Max Range: 0.4411. Min Range: -0.4411

Max Act: 0.3011. Min Act: -0.0000

Data Index: 218947 (C4 (Web Text))

Max Activating Token Index: 374

Text #14

Max Range: 0.4411. Min Range: -0.4411

Max Act: 0.3041. Min Act: -0.0000

Data Index: 123074 (C4 (Web Text))

Max Activating Token Index: 244

Text #15

Max Range: 0.4411. Min Range: -0.4411

Max Act: 0.3035. Min Act: -0.0001

Data Index: 212345 (C4 (Web Text))

Max Activating Token Index: 101

Text #16

Max Range: 0.4411. Min Range: -0.4411

Max Act: 0.2999. Min Act: -0.0000

Data Index: 1096645 (C4 (Web Text))

Max Activating Token Index: 243

Text #17

Max Range: 0.4411. Min Range: -0.4411

Max Act: 0.2978. Min Act: -0.0000

Data Index: 1264466 (C4 (Web Text))

Max Activating Token Index: 102

Text #18

Max Range: 0.4411. Min Range: -0.4411

Max Act: 0.2989. Min Act: -0.0001

Data Index: 1319991 (C4 (Web Text))

Max Activating Token Index: 826

Text #19

Max Range: 0.4411. Min Range: -0.4411

Max Act: 0.2873. Min Act: -0.0000

Data Index: 578870 (C4 (Web Text))

Max Activating Token Index: 125